Notice: Wiley Online Library will be unavailable on Saturday 27th February from 09:00-14:00 GMT / 04:00-09:00 EST / 17:00-22:00 SGT for essential maintenance. Apologies for the inconvenience.
Potential conflict of interest: Nothing to report.
We previously reported that liver natural killer (NK) and NKT cells play a critical role in mouse model of acetaminophen (APAP)-induced liver injury by producing interferon gamma (IFN-γ) and modulating chemokine production and subsequent recruitment of neutrophils into the liver. In this report, we examined the role of neutrophils in the progression of APAP hepatotoxicity. C57BL/6 mice were given an intraperitoneal toxic dose of APAP (500 mg/kg), which caused severe acute liver injury characterized by significant elevation of serum ALT, centrilobular hepatic necrosis, and increased hepatic inflammatory cell accumulation. Flow cytometric analysis of isolated hepatic leukocytes demonstrated that the major fraction of increased hepatic leukocytes at 6 and 24 hours after APAP was neutrophils (Mac-1+Gr-1+). Depletion of neutrophils by in vivo treatment with anti-Gr-1 antibody (RB6-8C5) significantly protected mice against APAP-induced liver injury, as evidenced by markedly reduced serum ALT levels, centrilobular hepatic necrosis, and improved mouse survival. The protection was associated with decreased FasL-expressing cells, cytotoxicity against hepatocytes, and respiratory burst in hepatic leukocytes. In intracellular adhesion molecule (ICAM)-1–deficient mice, APAP caused markedly reduced liver injury when compared with wild-type mice. The marked protection in ICAM-1–deficient mice was associated with decreased accumulation of neutrophils in the liver. Hepatic GSH depletion and APAP-adducts showed no differences among the antibody-treated, ICAM-1–deficient, and normal mice. In conclusion, accumulated neutrophils in the liver contribute to the progression and severity of APAP-induced liver injury. (HEPATOLOGY 2006;43:1220–1230.)
If you can't find a tool you're looking for, please click the link at the top of the page to "Go to old article view". Alternatively, view our Knowledge Base articles for additional help. Your feedback is important to us, so please let us know if you have comments or ideas for improvement.
Acetaminophen (APAP) is a widely used over-the-counter analgesic drug that is usually safe when used in therapeutic doses. In overdose, however, it can lead to acute liver failure characterized by centrilobular hepatic necrosis in experimental animals and in humans. APAP hepatotoxicity is the most common cause of acute liver failure in the United States1 and is associated with a significant number of deaths. The toxic response to APAP is thought to be initiated by a highly electrophilic intermediate, N-acetyl-p-benzoquinone-imine (NAPQI), generated by hepatic CYP2E1.2–4 Excessive NAPQI depletes hepatic glutathione (GSH) and covalently binds to cellular macromolecules, resulting in oxidative stress reactions, dysfunction of mitochondria, and DNA damage.4 These initiation events ultimately may result in direct damage to hepatocytes, leading to cell death of hepatocytes.
In addition to these initiation events occurring intracellularly in parenchymal hepatocytes after APAP exposure, growing evidence indicates that the liver innate immune system participates in the progression of liver injury.5–11 The inflammatory mediators such as cytokines, chemokines, reactive oxygen, and nitrogen species released by Kupffer cells/macrophages have been implicated in APAP hepatotoxicity.5–10 We have previously reported that natural killer (NK) and NKT cells, a major component of liver innate immune system, play a critical role in the severity and progression of APAP-induced liver injury by secreting cytokine interferon gamma (IFN-γ), modulating chemokine production and recruitment of inflammatory cells into the liver.11 Our previous study also identified neutrophils as a major fraction of inflammatory cells in the liver from APAP-challenged mice.11 Depletion of NK and NKT cells inhibited liver injury in association with markedly reduced accumulation of neutrophils in the liver,11 which prompted us to investigate the role of neutrophils in the progression of APAP hepatotoxicity. Neutrophil recruitment and activation have been implicated in a variety of liver injury models, such as alcoholic hepatitis,12 ischemia/reperfusion injury of liver,13 endotoxic shock,14 and concanavalin A–induced liver injury.15 However, the role of neutrophils in APAP-induced liver injury has been controversial,11, 16, 17 although the numbers of neutrophils were reported to be increased in the liver from APAP-challenged animals.10, 11, 16, 17 The decrease in neutrophil infiltration that we previously observed in NK/NKT cell–depleted mice and IFN-γ null mice could reflect a downstream role of neutrophils in the liver injury or could reflect the diminished need to resolve necrotic debris in protected livers. Therefore, in the current study, we directly investigated the role of neutrophils in liver injury in APAP-treated mice by in vivo depleting neutrophils with anti-Gr-1 antibody as well as assessing the influence of deletion of intracellular adhesion molecule (ICAM)-1, which is critical for neutrophil infiltration.
Pathogen-free male C57BL/6 (B6) and B6 ICAM-1–deficient mice (ICAM-1KO), 6 to 8 weeks of age, were obtained from the Jackson Laboratory (Bar Harbor, ME). All animals were fasted overnight before APAP (Sigma-Aldrich, St. Louis, MO) treatment. APAP was dissolved in warmed phosphate-buffered saline (PBS), and the animals were treated intraperitoneally with APAP at the dose of 500 mg/kg or PBS as controls.
To deplete neutrophils, mice were injected intraperitoneally with 250 and 100 μg per mouse of anti-Gr-1 monoclonal antibody (mAb) (RB6-8C5) (BD Biosciences, San Diego, CA) at 24 and 4 hours before APAP challenge, respectively. Previous studies18, 19 and our pilot experiments have confirmed that anti-Gr-1 treatment specifically depletes both peripheral and liver leukocytes with an efficiency of greater than 95% as determined by peripheral blood smear staining and flow cytometric analysis of isolated liver leukocytes. Control rat IgG2b (BD Biosciences) for anti-Gr-1 treatment were injected at an equivalent dose and schedule.
For assay of serum liver enzyme, ALT, mice were anesthetized and bled from the retroorbital venous plexus at the indicated times.11 Serum ALT levels were measured at the USC Pathology Reference Laboratory. Animal experiments were carried out in accordance with guidelines from the USC Institutional Animal Care and Use Committee.
Preparation of Hepatic Leukocytes and Flow Cytometric Analysis.
Liver leukocytes were isolated as previously described.11, 20–22 In brief, liver tissues were passed through a 200-gauge stainless steel mesh in Hank's balanced salt solution. The cell suspension was centrifuged at 500g for 5 minutes, and the resulting cell pellet from one liver was re-suspended in 15 mL 35% Percoll containing 100 U/mL heparin. The cell suspension was centrifuged at 500g for 15 minutes at room temperature, and the cell pellet containing leukocytes was harvested and re-suspended in 5 mL red blood cell lysis solution (155 mmol/L NH4CL, 10 mmol/L KHCO3, 1 mmol/L EDTA, 170 mmol/L Tris, pH 7.3). After incubation for 5 minutes on ice, cells were washed twice in RPMI 1640 containing 5% fetal bovine serum.
For flow cytometric analysis, 106 cells were first incubated for 10 minutes with Fcγ receptor blocker (CD16/32, 2.4G2; BD Biosciences, San Diego, CA), then stained with mAbs at a concentration of 1 μg per 100 μL PBS containing 0.2% bovine serum albumin (Boehringer Mannheim, Indianapolis, IN) and 0.02% sodium azide for 30 minutes on ice.21 The following antibodies were used: fluorescein isothiocyanate (FITC)-conjugated anti-CD3 (145-2C11), anti-CD11b (Mac-1; M70) and phycoerythrin (PE)-conjugated anti-NK1.1 (PK136), anti-Gr-1 (Ly-6G, 1A8), all purchased from BD Biosciences (San Diego, CA). FITC-conjugated anti-F4/80 mAb was purchased from Serotec Inc (Raleigh, NC). For detection of FasL expression, cells were first stained with biotinylated anti-mouse FasL (BD Biosciences) and then stained with Cy-chrome conjugated streptavidin. Flow cytometric analysis was performed on a FACSCalibur (Becton Dickinson, Mountain View, CA). The absolute number of T cells, NK cells, NKT cells, neutrophils, and macrophages per liver was calculated by multiplying the percentage of each population from two parameter flow cytometry data by the total number of isolated leukocytes per liver.11, 21, 22
Histology and Measurement of Hepatic Glutathione.
For histology, liver tissue was fixed in 10% neutral buffered formalin and sections stained with hematoxylin-eosin (HE)to determine morphological changes. HE-stained liver sections were examined under low-power (100×) microscopy and necrosis graded using a system23 previously described: 0, no lesion present; 1/2, individual necrotic cells seen at the first cell layer adjacent to the central vein, and hyaline degeneration present; 1, necrotic cells extending two or three cell layers from the central veins; 2, necrotic cells extending three to six cell layers from the central veins, but limited in peripheral distribution; 3, the same as 2, but with necrosis extending from one central vein to another; 4, more severe than 3, with extensive centrilobular necrosis throughout the section. An overall score was computed for each liver section based on assessment of five lobules.
Hepatic total GSH was measured by the DTNB-glutathione reductase recycling assay (Tietze's method) as previously described.11, 24 The hepatic GSH concentrations were adjusted according to original wet weight of liver tissue (nmol/mg liver tissue).
Preparation of Liver Extracts and Western Blot Analysis.
Liver tissues were homogenized in lysis buffer (30 mmol/L Tris pH 7.5, 1% Nonidet P-40, 150 mmol/L sodium chloride, cocktail of proteinase inhibitors) at 4°C, and centrifuged at 12,000 rpm at 4°C for 15 minutes. The supernatants were collected and protein concentration was measured with the Bradford assay (Bio-Rad Laboratories Inc.). Aliquots of liver extracts containing 40 μg protein were mixed in 2× Laemmli loading buffer, boiled for 5 minutes, and then subjected to 12% SDS-PAGE. After electrophoresis, proteins were transferred to a PVDF Hybond-P membrane (Amersham Biosciences, Buckinghamshire, UK). Membranes were first probed for 90 minutes at room temperature using a specific polyclonal rabbit anti-APAP serum (provided by Dr. Jack A. Hinson, University of Arkansas for Medical Sciences), at 1:5,000 dilution, and then incubated with horseradish peroxidase–conjugated goat anti-rabbit IgG secondary antibody (BD PharMingen, San Diego, CA) diluted 1:5,000 for 60 minutes at room temperature. Blots were developed by the enhanced chemiluminescence system (Amersham Bioscience) according to the manufacturer's instructions.
Hepatocyte Isolation and Cytotoxicity.
Primary mouse hepatocytes were isolated by the procedure of collagenase perfusion as previously described.25 The cytolytic activity of hepatic leukocytes (effector cells) against hepatocytes (target cells) was determined using the aspartate transaminase (AST) release assay.26, 27 Briefly, mouse hepatocytes from non-treated normal B6 mice were distributed into rat tail collagen–coated 48-well tissue culture plates at a final concentration of 2 × 104 cells/well, and cultured for 3 hours at 37°C before replacing the supernatants with fresh culture medium. Then isolated hepatic leukocytes (4 × 105 cells/well) were added at an effector-to-target (E/T) ratio of 20:1, which was determined to be optimal, considering both killing activity and the yield of isolated hepatic leukocytes, in our preliminary titration experiments for different E/T ratios from 5:1 to 100:1. After 6 hours' co-culture at 37°C incubator, supernatant from each well was harvested, and the AST activity in the supernatant was measured using an AST kit purchased from Sigma-Aldrich. To assess whether cell contact is required for cytotoxicity, hepatic leukocytes were separated from hepatocytes by placing the hepatic leukocytes on the membrane surface of a transwell insert (Corning Inc., Corning, NY) in the top chamber and hepatocytes on the bottom of the tissue culture plate. The specific cytotoxicity was calculated as [(experimental AST − spontaneous AST)/(total AST − spontaneous AST)] × 100%.26, 27 Total AST activity was determined by lysing the cells with 0.05% Triton X-100.
Measurement of Respiratory Burst From Hepatic Leukocytes.
Respiratory burst from freshly isolated hepatic leukocytes (5 × 105 cells/sample) was measured after phorbol myristate acetate (PMA) (10 ng) stimulation using horseradish peroxidase with phenylhydroxylacetate in the presence of superoxide dismutase by a spectrofluorometer as previously described.28 Superoxide and H2O2 release from leukocytes during respiratory burst was measured as H2O2 because of the presence of superoxide dismutase.
For mouse survival studies, the log-rank test was used to determine significance. Histological scores were expressed as median and evaluated by Kruskal-Wallis nonparametric analysis. For all other studies, data are expressed as mean ± SEM. Group comparisons were performed using Student's t test or analysis of variance. Differences were considered statistically significant at P < .05.
Increased Infiltration of Neutrophils in the Liver From APAP-Treated Mice.
Our previous studies have shown that a toxic dose of APAP caused severe acute liver injury accompanied by increased accumulation of hepatic neutrophils.11 In Fig. 1A, we confirmed that in normal B6 mice intraperitoneal injection of APAP (500 mg/kg) induced significant liver injury indicated by a marked increase in serum ALT levels at 6 hours, peaking at approximately 24 hours. A less marked increase in ALT was first observed at 4 hours. Based on two-parameter flow cytometric analysis of cell type markers for T cells (CD3+ NK1.1−), NK cells (NK1.1+CD3−), NKT cells (NK1.1+CD3+), neutrophils (Gr-1+Mac-1+), and macrophages (F4/80+), we found that neutrophils accounted for the major fraction of accumulated hepatic leukocytes in APAP-treated mice. As shown in Fig. 1B, compared with non-treated control mice, APAP induced moderate increases of NK, NKT, and macrophages at 6 and 24 hours after APAP treatment. However, neutrophils accounted for the major increase in the hepatic leukocytes from 0.32 ± 0.15 × 106 cells per liver in non-treated control mice to 2.8 ± 1.2 × 106 cells at 6 hours and 3.9 ± 1.4 × 106 at 24 hours in APAP-treated mice (P < .05 vs. non-treated control mice, respectively).
Effect of Depletion of Neutrophils on APAP-Induced Liver Injury.
To determine whether the infiltrated hepatic neutrophils are involved in the development of APAP-induced liver injury, a group of B6 mice were depleted of neutrophils by anti-Gr-1 mAb intraperitoneal injection at 24 hours (250 μg per mouse) and 4 hours (100 μg per mouse) before APAP challenge (500 mg/kg). As shown in Fig. 2A, anti-Gr-1 treatment markedly decreased the serum ALT levels at 6 and 24 hours when compared with that of APAP-treated control B6 mice (500 mg/kg) (P < .05). In control antibody (IgG2b)–treated mice, no protection against APAP-induced liver injury occurred (Fig. 2A). Furthermore, histological analysis of liver sections at 24 hours after APAP challenge showed a considerably reduced area of centrilobular hepatic necrosis with a grade of 2.0 (n = 10) in neutrophil-depleted mice compared with APAP-treated control mice, in which APAP caused massive centrilobular hepatic necrosis with a grade of 3.8 (n = 12) (P < .05). A representative HE staining of liver sections is shown in Fig. 2B. In addition, as shown in Fig. 2C, depletion of neutrophils significantly protected from APAP-related mouse death. APAP 500 mg/kg caused approximately 42% mortality (10 of 24 mice) by 3 days in normal B6 mice, whereas only approximately 5% mortality was observed in the anti-Gr-1–treated mice (1 of 19 mice) (P < .05).
Effect of Depletion of Neutrophils on APAP-Induced Inflammatory Cell Accumulation in the Liver.
To examine whether the protection by depletion of neutrophils is associated with changes of innate immune cells in the liver, we determined the effect of neutrophil depletion on the leukocyte accumulation in the liver. To examine this, hepatic leukocytes were isolated from mouse liver at 24 hours after APAP treatment. The total numbers of hepatic leukocytes were significantly increased in APAP-treated mice (7.6 ± 0.8 × 106 cells per liver at 24 hours, n = 6) compared with non-treated control B6 mice (2.6 ± 0.5 × 106, n = 4). Anti-Gr-1 treatment reduced APAP-induced leukocyte accumulations by 48% at 24 hours after APAP (3.9 ± 0.5 × 106, n = 6) (P < .05). Next we determined the cell types of liver leukocytes by two-parameter flow cytometric analysis. Representative data at 24 hours after APAP challenge are shown in Fig. 3A. The major increase was in the neutrophil fraction (Gr-1+Mac-1+) from 10% in non-treated control mice to 49% in APAP-treated mice (Fig. 3A, circled populations in middle panels). The efficiency of in vivo depletion of neutrophils by anti-Gr-1 treatment was confirmed. Less than 1% of hepatic leukocytes were neutrophils in anti-Gr-1–treated mice (Fig. 3A, middle panels). Compared with APAP-treated mice, there were no significant changes in the fraction of NK cells (NK1.1+CD3−), NKT cells (NK1.1+CD3+), T cells (CD3+ NK1.1−) and macrophages (F4/80+). Based on the flow cytometric analysis, the numbers of each cell type per liver were calculated and are shown in Fig. 3B. Compared with APAP-treated mice, anti-Gr-1 pretreatment decreased the numbers of neutrophil from 3.6 ± 0.9 × 106 cells per liver in APAP-treated mice to 0.17 ± 0.11 × 106 cells per liver, which was even below the neutrophil numbers in non-treated normal controls (0.32 ± 0.17 × 106 cells per liver). However, anti-Gr-1 had only a modest inhibitory effect on the infiltration of other inflammatory cells in the liver (Fig. 3B). Therefore, anti-Gr-1 mAb treatment specifically eliminated the neutrophil recruitment into the liver, which was the major cell population of inflammatory cells in the liver of APAP-treated mice.
APAP-Induced Liver Injury in ICAM-1–Deficient Mice.
ICAM-1 is an important adhesion molecule in neutrophil recruitment and extravasation through endothelial cells in inflammatory sites and has been implicated in liver injury models such as endotoxic shock and ischemia/reperfusion–induced liver injury.14, 29, 30 Upregulated expression of ICAM-1 in the liver has been reported in APAP hepatotoxicity.10 To examine whether ICAM-1 plays a role in the neutrophil accumulation in APAP-treated mice, experiments were carried out in ICAM-1–deficient mice. As shown in Fig. 4A, the serum ALT levels at 6 and 24 hours after APAP challenge were significantly lower in ICAM-1 null mice than that in wild-type control mice (P < .05). Significant protection against APAP-induced histological liver injury was also observed in ICAM-1 null mice with a histology grade of 2.4 (n = 8) compared with a grade of 3.6 (n = 10) (P < .05) in APAP-treated control mice. Furthermore, we examined the inflammatory cell infiltration in APAP-treated ICAM-1 null mice. When compared with APAP-treated wild-type mice (6.5 ± 0.7 × 106 cells per mouse liver at 24 hours, n = 4), the numbers of total hepatic leukocytes per mouse liver were significantly reduced in APAP-treated ICAM-1 null mice (3.2 ± 0.5 × 106 cells per mouse liver at 24 hours after APAP treatment, n = 4, P < .05)(Fig. 4B). The subpopulation examination by flow cytometric analysis demonstrated that the significant reduction of hepatic leukocytes in ICAM-1 null mice was in the fraction of neutrophils (Mac-1+Gr-1+) (Fig. 4B). These results indicate that ICAM-1–dependent neutrophil recruitment in the liver plays a role in the development of APAP-induced liver injury.
Cytotoxicity of Hepatic Leukocytes Against Primary Murine Hepatocytes.
Neutrophils have the capability to produce and release a variety of cytotoxic agents such as reactive oxygen and nitrogen species, inflammatory cytokines, bioactive lipids, and hydrolytic proteinases.31 Several in vitro studies have clearly demonstrated that activated neutrophils could cause hepatocellular injury.32, 33 To determine whether hepatic neutrophils isolated from APAP-treated mice are capable of killing primary hepatocytes, we co-cultured hepatic leukocytes with primary hepatocytes from nontreated normal B6 mice for 6 hours; the cytotoxicity was calculated by measuring the released AST values from the co-culture medium. As shown in Fig. 5, cytotoxicity against hepatocytes was significantly increased in co-culture with hepatic leukocytes from APAP-treated mice compared with that from non-treated mice. Interestingly, APAP-induced hepatic leukocyte cytotoxicity against hepatocytes was significantly decreased in anti-Gr-1–pretreated mice compared with that in APAP-treated control mice (P < .05), suggesting the important role of direct neutrophil cytotoxicity against hepatocytes. To determine whether hepatic leukocytes from APAP-treated mice kill hepatocytes through diffusible mediators or direct cell contact is required, we co-cultured hepatic leukocytes with hepatocytes in a no cell-contact fashion through a transwell co-culture system by placing hepatic leukocytes on the top chamber of transwell inserts and hepatocytes on the bottom of tissue culture plate. As shown in Fig. 5B, cytotoxicity under transwell co-culture accounted for about 65% of the total cytotoxicity, suggesting the diffusible mediators released by activated neutrophils may be dominant for lysis of hepatocytes.
Respiratory Burst in Hepatic Leukocytes.
Oxidative stress caused by the generation of reactive oxygen species is implicated in APAP-induced liver injury.4, 6, 9 Activated phagocytes undergo respiratory burst to generate superoxide and hydrogen peroxide, which are important in killing invading organisms, and are also capable of inflicting damage to surrounding tissues. Respiratory burst was measured in isolated hepatic leukocytes stimulated by PMA to examine whether neutrophils contribute to oxidative stress in the liver. As shown in Fig. 6, reactive oxygen species generation was significantly increased from freshly isolated hepatic leukocytes of APAP-treated mice. However, the hepatic leukocytes from neutrophil-depleted mice generated much less reactive oxygen species compared with that in APAP-treated control mice (Fig. 6) (P < .05).
Effect of APAP and Depletion of Neutrophils on FasL Expression in the Liver.
In our previous study, we have shown that Fas/FasL pathway participates in the development of APAP-induced liver injury.11 Because neutrophils, in addition to NK, NKT cells, and macrophages, also might use the FasL effector system to modulate liver injury, we examined the effect of neutrophil depletion on the FasL expression in the liver. By flow cytometric analysis of isolated liver leukocytes, we found that the fraction of FasL-expressing cells among liver leukocytes was increased in APAP-challenged mice (39% ± 8%) compared with that of non-treated control mice (9% ± 6%) (Fig. 7A), and that depletion of neutrophils significantly reduced the FasL-expressing cells in the liver (22% ± 7%) in neutrophil-depleted mice (P < .05 vs. APAP-treated control mice) (Fig. 7A). Representative data of flow cytometry analysis is shown in Fig. 7B. These results indicate that neutrophils are an important cell population of hepatic inflammatory cells that express increased FasL on their cell surface.
APAP Metabolism in the Control, Anti-Gr-1–Treated, and ICAM-1–Deficient Mice: Effects on GSH Depletion and Covalent Adducts.
To rule out the possibility that the inhibition of APAP-induced liver injury by anti-Gr-1 treatment and the resistance of ICAM-1–deficient mice were due to alterations of APAP metabolism, hepatic GSH concentrations were measured at various times after APAP challenge (Fig. 8A). There was no difference in basal levels, and the depletion of GSH at 1 and 4 hours after APAP challenge or the recovery at 24 hours in control mice versus anti-Gr-1–treated mice and ICAM-1-null mice. We examined APAP covalent binding to liver proteins by Western blotting with anti-APAP antiserum. As shown in Fig. 8B, in the liver protein extracts from APAP-treated mice, multiple bands of covalent binding with the molecular masses between 30 and 100 kd were detected. However, no qualitative differences of covalent binding to individual proteins in APAP treated wild-type mice versus anti-Gr-1–treated mice and ICAM-1-null mice were seen (Fig. 8B).
Neutrophils are an essential component of the host innate immunity system. However, excessive response of neutrophils contributes to the development and perpetuation of inflammatory response. Neutrophils have been implicated in several liver injury models such as alcoholic hepatitis,12 ischemia/reperfusion injury of liver,13 endotoxic shock,14 adenovirus-induced liver injury,34 obstructive cholestasis,35 and concanavalin A-induced liver injury.15 In this study, by in vivo specific depletion of neutrophils using a well-established neutralizing antibody, anti-Gr-1, we have demonstrated the important role of neutrophils in the progression of APAP-induced liver injury. After challenge with a toxic dose of APAP, neutrophil-depleted C57BL/6 mice exhibited markedly reduced liver injury as evidenced by significant decreases of serum ALT and histological liver injury as well as significantly improved mouse survival when compared with APAP-treated control mice. Another finding in the current study is that ICAM-1–deficient mice were resistant to APAP-induced liver injury compared with wild-type mice (Fig. 4). The partial protection in ICAM-1–deficient mice was associated with reduced neutrophil accumulation in the liver. These results suggest that neutrophil recruitment and accumulation in the liver after APAP challenge are at least partially ICAM-1 dependent.
In support of our results showing the important role of neutrophils in APAP hepatotoxicity, neutrophil accumulation in the liver has been demonstrated by histochemical immunolabeling and by flow-cytometric analysis of isolated hepatic leukocytes in APAP-treated animals.10, 11, 16, 17 Furthermore, a report from Smith et al.16 showed protection against APAP-induced liver injury by in vivo injection of rabbit anti-neutrophil antiserum in a rat model. However, the role of neutrophils in APAP-induced liver injury has been controversial. A study by Lawson et al.17 has shown that the neutrophils appear to be recruited into the liver for removal of cell debris and do not directly contribute to the liver injury process in APAP hepatotoxicity, although the same group has demonstrated the critical role of accumulated leukocytes in other acute liver injury models such as endotoxic shock,14 ischemia/reperfusion,13 and bile-duct ligation.35 The discrepancy between their study and the current study may be explained by different approaches applied to examine the role of neutrophils. In their study,17 neutrophils were treated with an anti-CD18 antibody, which did not deplete neutrophils but had been shown in a previous study to decrease by 50% the accumulation of neutrophils in the liver.14 Conceivably, the remaining neutrophils may have been enough to aggravate APAP hepatotoxicity, because even the almost complete neutrophil depletion from the liver obtained in the current study afforded only partial protection against APAP-mediated toxicity to the liver. Another possible explanation could involve the Fas system. Our previous study11 has shown that the Fas/Fas ligand system plays a role in APAP-induced hepatitis, and the current study shows that neutrophil depletion decreases the number of Fas ligand–expressing hepatic leukocytes (Fig. 7). Whereas neutrophil depletion will obviously prevent neutrophil-expressed Fas ligand from causing liver injury, in contrast FasL/Fas pathway might be still functional when animals are only exposed to a neutrophil-inhibiting, anti-CD18 antibody.17 Our study showed an increase of neutrophil accumulation in the liver started at as early as 4 to 6 hours after APAP challenge, almost parallel to liver injury. Certainly the early infiltration of leukocytes that accompanies the appearance of parenchymal injury plays a role in the development of injury, whereas the later further increase in inflammatory cell infiltration could be important in “clean-up.”
Although anti-Gr-1 antibody has been demonstrated to specifically bind to and selectively deplete mature murine neutrophils in vivo in a variety of disease models including liver injury,15, 18, 19 the fate of the depleted neutrophils is not clear. Antibody-tagged neutrophils may accumulate in the liver and be removed by Kupffer cells. This may first activate and eventually overload and then inactivate Kupffer cells with a potential impact on APAP hepatotoxicity. To address this, we have performed experiments by using lipopolysaccharide (LPS)-induced acute liver injury model, which is mainly mediated by activation of Kupffer cells and production of tumor necrosis factor alpha.36, 37 We compared the sensitivity to a low non-lethal dose of LPS in anti-Gr-1 antibody pre-treated mice with that in control mice. Our results showed that LPS injection (1 mg/kg, intraperitoneally) caused a significant increase of serum ALT levels (at 3 and 6 hours after LPS) and up-regulated mRNA expression of tumor necrosis factor alpha and CD14 (3 hours after LPS) in the liver; however, there is no difference between LPS-challenged control and neutrophil-depleted mice (See Supplemental Fig. 1 at the HEPATOLOGY website: http://interscience.wiley.com/jpages/0270-9139/suppmat/index.html), indicating that LPS-induced activation of Kupffer cells is comparable in both groups. Together with our data showing no neutrophil accumulation in the liver after anti-Gr-1 treatment (Fig. 3), depletion of neutrophils by ant-Gr-1 treatment is unlikely to have a significant effect on Kupffer cell function.
Within the liver, activated neutrophils may act as effector cells through cytotoxicity, leading to hepatocyte death. Our in vitro studies on effector functions of neutrophils, in terms of cytotoxicity against primary mouse hepatocytes, generation of reactive oxygen intermediates, and FasL expression of hepatic leukocytes, offer a possible explanation for the in vivo observation of inhibition of APAP-induced liver injury in neutrophil-depleted mice. Our results suggest that accumulated neutrophils in the liver are important effector cells mediating hepatocyte death, and the cytotoxicity against hepatocytes appears mainly due to diffusible inflammatory mediators released from activated hepatic leukocytes (Fig. 5). Increased reactive oxygen formation has been implicated in APAP hepatotoxicity.6, 9, 25 Our results indicate that neutrophils are important cellular sources of reactive oxygen intermediates in the liver (Fig. 6), which might be involved in mediating hepatocyte death.
We and others have shown that Fas/FasL pathways are involved in the progression of APAP-induced liver injury.11, 38 Activated liver innate immune cells, including NK/NKT cells, macrophages, and neutrophils, can express FasL.11, 39–41 These effector cells may directly kill Fas-expressing hepatocytes.39 Alternatively, Fas-FasL crosslinking between innate immune cells and target cells such as hepatocytes, or even among the innate immune cells themselves within the liver, may have a pro-inflamamtory activity by inducing inflammatory cytokines/chemokines,11, 42–45 thereby aggravating APAP-induced liver injury. In the current study, we observed a marked reduction of FasL-expressing hepatic leukocytes in neutrophil-depleted mice compared with that in APAP-treated control mice, suggesting that neutrophils are important FasL-expressing cells in the liver and may contribute to APAP-induced liver injury through Fas/FasL pathways.
In summary, this study provides direct evidence showing that accumulated hepatic neutrophils contribute to the progression and severity of APAP-induced liver injury. Within the liver, recruited and activated neutrophils may mediate liver injury through direct cytotoxicity. Thus, interventions that inhibit hepatic infiltration of neutrophils may be useful in settings of APAP hepatotoxicity to ameliorate the progression and severity of liver injury.